Psychopharmacology

, Volume 172, Issue 4, pp 375–383

AR-R 17779 improves social recognition in rats by activation of nicotinic α7 receptors

Authors

  • Marja van Kampen
    • CNS ResearchBayer AG
    • ICON Clinical Research GmbHHeinrich-Hertz-Strasse 26
  • Karin Selbach
    • CNS ResearchBayer AG
  • Renate Schneider
    • CNS ResearchBayer AG
  • Elleonore Schiegel
    • CNS ResearchBayer AG
  • Frank Boess
    • CNS ResearchBayer AG
    • CNS ResearchBayer AG
    • CNS NeurobehaviorRoche Pharmaceuticals
Original Investigation

DOI: 10.1007/s00213-003-1668-7

Cite this article as:
van Kampen, M., Selbach, K., Schneider, R. et al. Psychopharmacology (2004) 172: 375. doi:10.1007/s00213-003-1668-7

Abstract

Rationale

Nicotine and agonists at α4β2 and α7 nicotinic acetylcholine receptors (nAChRs) improve learning and memory. The α7-nAChR subtype is of special interest, since it appears to play no role in the abuse liability of nicotine.

Objectives and methods

To further investigate the role of the α7-nAChR in learning and memory, the effects of the specific α7-nAChR agonist AR-R17779 on cognition were measured in the rat social recognition test (SRT) and the effect of the α7-nAChR antagonist methyllycaconitine (MLA) was studied. The SRT and a scopolamine-induced deficit version were validated with the acetylcholinesterase inhibitor metrifonate. Social memory was measured by the ability of an adult rat to recognize a juvenile rat after a delay. The difference in social interaction time (SIT) was measured between two encounters. The difference in SIT is expressed as percent reduction in social interaction time (%RSIT).

Results

Metrifonate (10 and 30 mg/kg PO) increased %RSIT in a behaviorally specific manner, employing a 24-h interval and reversed the scopolamine-induced deficit at a retention time of 15 min. Likewise, AR-R17779 increased %RSIT in unimpaired animals (1, 3, 10 and 30 mg/kg SC) employing a 24-h retention interval, and reversed the scopolamine-induced deficit (0.3 and 1 mg/kg SC) after a 15-min retention interval. The effects of AR-R17779 (1 mg/kg SC) in unimpaired animals were reversed by MLA (10 µg ICV), which induced a decrease of %RSI at a 15-min retention interval when given alone.

Conclusions

AR-R17779 increased social recognition memory by activation of α7-nAChRs, suggesting that α7-nAChR agonists possess cognitive-enhancing properties.

Keywords

α7-nicotinic receptorsSocial memoryCognitionAcetylcholinesterase inhibitorsScopolamine

Introduction

Nicotine enhances cognitive functions such as attention, learning and memory, and retention in both animals and humans, through activation of brain nicotinic acetylcholine receptors (nAChRs) (Levin and Simon 1998; Levin et al. 1999). These receptors are homo- or heteropentameric ligand-gated ion channels. They are formed by two subunit classes (α and β) and different isoforms of these subunits have been identified (α2–α10; β2–β4; for review, see Paterson and Nordberg 2000; Elgoyhen et al. 2001). The most common nicotinic receptors found in the brain are the α7-nAChR and the α4β2-nAChR. Evidence from neuroanatomical, electrophysiological, and behavioral studies supports a role for both of these receptor subtypes in processes of learning and memory.

Studies using [125I]-α-bungarotoxin and [3H]-cytisine to label α7-nAChRs and α4β2-nAChRs, respectively, have identified high densities of these receptors in the hippocampus, a brain area which plays an important role in learning and memory processes (Paterson and Nordberg 2000). These hippocampal receptors are involved in long-term potentiation (LTP), proposed as a potential cellular mechanism for learning and memory (Bliss and Collingridge 1993). The underlying mechanism may involve GABA-ergic inhibition of interneurons containing nAChRs. Subsequent disinhibition of pyramidal dendrites might lead to LTP-like synaptic activation (Albuquerque et al. 2000). The α4β2-nAChR agonist A-85380, the α7-nAChR agonist AR-R 17779 (Gordon et al. 1998), and the α7- (and weak α4β2-) nAChR agonist DMXB (GTS-21) (Hunter et al. 1994; de Fiebre et al. 1995) facilitate the induction of hippocampal LTP in rats (Hunter et al. 1994; Chen et al. 2000; Fuji and Sumikawa 2001; however, see also Fujii and Sumikawa 2000). Co-application of the α7-nAChR antagonist methyllycaconitine (MLA) inhibited the effects of AR-R 17779, whereas MLA alone had no effects on LTP (Chen et al. 2000).

In line with the hypothesis that LTP may be a cellular substrate for learning and memory, α7- and α4β2-nAChR agonists possess cognitive-enhancing properties (for reviews, see Buccafusco et al. 1995; Terry et al. 1997; Kem 2000). As the β2-subunit is involved in the addictive effects of nicotine (Picciotto et al. 1998), and α7-nAChRs are apparently not involved in the rewarding effects of nicotine (Brioni et al. 1996; Grottick et al. 2000), the likelihood that an α7-nAChR agonist possesses no abuse potential adds to the attractiveness of such a compound as a cognitive enhancer. GTS-21 displays cognitive-enhancing properties in various animal models: 1) it enhances inhibitory avoidance responding in rats (Meyer et al. 1994), 2) it facilitates performance in long-delay trials of a delayed-matching-to-sample test in monkeys (Briggs 1997), 3) it increases acquisition of eye blink conditioning in rabbits (Woodruff-Pak et al. 1994), 4) it improves performance in the Lashley III maze and one-way active avoidance, and 5) it increases general learning and reference memory in the 17-arm radial maze in rats (Arendash et al. 1995). GTS-21 is not selective for the human α7-nAChR (Briggs et al. 1997), but its metabolite 4-OH-GTS-21 is a selective partial agonist at both the human and rat α7-nAChR (Meyer et al. 1998). GTS-21 may be a pro-drug whose efficacy in primate behavioral tests is the result of its active metabolite (Meyer et al. 1998). Also, in humans, GTS-21 has a favourable effect on cognitive function, especially on attention, working memory and episodic secondary memory (Kitagawa et al. 2003). The more selective α7-nAChR agonist, AR-R 17779, improved long-term win-shift acquisition in an eight-arm radial maze (Levin et al. 1999). Infusion of MLA into the hippocampus impaired working memory in a similar radial arm maze task (Felix and Levin 1997). However, AR-R 17779 failed to improve performance in a five-choice serial reaction time task (Grottick and Higgins 2000; Hahn et al. 2001). Together, these results indicate a role for α7-nAChRs in learning and memory, and, possibly, attention processes.

To further elucidate the role of α7-nAChRs in learning and memory, the rat social recognition test, first described by Thor and Holloway (1982), was used. Rats have an innate interest in their conspecifics, and their olfactory discrimination capabilities can easily be used to measure the difference in interest between two encounters. Social recognition is measured by the difference in investigation by an adult rat of a juvenile rat between the first and the second encounter. This test is reported to be sensitive to acetylcholinesterase inhibitors (AChE-Is) (Winslow and Camacho 1995) and neurohypophyseal peptides (Dantzer et al. 1987; Everts and Koolhaas 1997). In mice, scopolamine induces a recognition deficit, which is reversed by AChE-Is (Winslow and Camacho 1995). The present study investigated the effect of the AChE-I metrifonate (Nordgren et al. 1981) and the α7-nAChR agonist AR-R 17779 in the social recognition test in unimpaired rats, and in a scopolamine deficit model. Metrifonate was chosen for its good tolerability in rodents and its robust effects in a wide range of cognition models (i.e. shuttle box active avoidance, Morris water maze, cone field, eyeblink conditioning, radial arm maze, object recognition, and passive avoidance) (van der Staay et al. 1996; Dachir et al. 1997; Itoh et al. 1997a, 1997b; Kronforst-Collins et al. 1997; Riekinnen et al. 1997a, 1997b; Scali et al. 1997; Prickaerts et al. 1999). These studies were performed in both unimpaired rats and impaired rats (i.e. aged animals or scopolamine treatment).

Material and methods

Animals

Male Wistar rats (Harlan-Winkelman, Borchen, Germany; 4 months old; weighing 380–460 g, and 5–6 weeks old; weighing 90–140 g) were housed in groups of three per cage (from polycarbonate, type IV for the 4-month-old rats and type III for the 5- to 6-week-old-rats) under a 12-h light-dark schedule (lights on at 7:00 a.m.). Food (Sniff, Soest, Germany) and water were available ad libitum. Ambient room temperature (22±1°C) and relative humidity (55±5%) were kept constant. Experimental protocols and conditions conformed to the local regulations on animal welfare and were in accordance with the Helsinki agreement.

General experimental procedure

Animals were adapted to laboratory housing conditions for 1 week prior to behavioral testing. One habituation session was performed under essentially similar conditions as the test session (see below).

Trial 1

Adult animals were individually housed 30 min before testing. An enclosure (63×41×40 cm; two aluminum side walls, one aluminum back wall, and one Plexiglas front), was put over the cage 4 min prior to testing with the lid of the cage removed. For testing, a juvenile was placed into the cage and the social investigation by the adult rat was measured cumulatively for 2 min by a trained observer. Sniffing and grooming of body parts, anogenital sniffing and close following were scored. Contact between an adult and a juvenile is not always equal to social investigation. Sometimes contacts are by accident and the adult is obviously not interested in the juvenile, as is seen by the nose of the adult rat, which is not pointed to the juvenile. Such contacts were not scored. During tests with retention times less than 2 h, animals remained singly housed. During tests with longer retention times, animals were put back in their home cage.

Trial 2

The retest was performed after a delay under essentially the same experimental conditions as trial 1. In the tests conducted with a familiar juvenile, each adult is exposed to the same juvenile already used during trial 1. In tests conducted with an unfamiliar juvenile, each adult is exposed to a juvenile they have not previously experienced, for verification of the behavioral specificity of a drug effect. The percent reduction in social investigation time (%RSIT) between trials 1 and 2 was used as an index of social memory.

Surgery

Rats were anaesthetized with a 90 mg/kg ketamine-12 mg/kg xylazine mixture (6 ml/kg, IP), placed in a stereotaxic apparatus, and implanted with a stainless steel cannula (23 gauge, 6.0 mm length) aimed at the lateral ventricle (interaural line coordinates for the tip of the cannula: F 8.2, L –1.4, H 6.5, according to Paxinos and Watson 1982). The tip of the cannula was advanced to 1 mm above the lateral ventricle. Two stainless steel holding screws (1.4×3.0 mm) were secured into the surface of the skull, and the skull around the cannula was covered with dental cement. A recovery period of at least 7 days was allowed before the social recognition test was performed. Histological verification of the coordinates was performed previously in a number of animals. These studies consistently confirmed the correct placement of the cannulae. Therefore, verification of the coordinates was confirmed in randomly selected experimental subjects. In experiment 6, MLA injections into the lateral ventricle were made with a Hamilton cannula (2 µl injection volume).

Drugs

Methyllycaconitine (MLA) and scopolamine were obtained from Research Biochemicals International/Sigma-Aldrich (Deisenhofen, Germany). AR-R 17779 {3-[2,4-dimethoxybenzylidene]anabaseine} and metrifonate were synthesized by Bayer Chemistry Departments (Leverkusen and Wuppertal, Germany). Metrifonate was dissolved in 5% sodium citrate (pH 5.5); all other compounds were dissolved in 0.9% NaCl.

Experimental procedure

For all experiments (1–6), the social recognition test was essentially performed as described in the general experimental procedure. The specific experimental conditions for the six experiments are depicted in Table 1.
Table 1

Summary of the specific test conditions for experiments 1–6

Experiment

Retention time

Juvenile

Treatment 1 (dose range in mg/kg)

Treatment 2 (dose range in mg/kg)

1

10 min to 24 h

Familiar

None

None

2

24 h

Familiar and novel

Metrifonate (3–30 and 30, PO)

None

3

15 min

Familiar

Scopolamine (0.01–0.1, SC)

None

Scopolamine (0.1, SC)

Metrifonate (10–30, PO)

4

24 h

Familiar and novel

AR-R 17779 (0.3–30 and 1.0, SC)

None

5

15 min

Familiar

AR-R 17779 (0.3–3, SC)

Scopolamine (0.1, SC)

6

24 h

Familiar

AR R 17779 (1, SC)

MLA (10 μg/rat, ICV)

15 min

Familiar

MLA (10 μg/rat, ICV)

None

In experiments aimed to detect drug-induced improvements of social recognition, we used a 24-h retention interval between the first and the second encounter with the same juvenile. In experiments aimed to detect a drug-induced reversal of a scopolamine-induced deficit in social recognition, we used a 15-min retention interval.

Experiment 1: establishment of a forgetting curve

Trial 2 was conducted at different retention times, ranging from 10 min to 24 h, with the adult animal being exposed to the familiar juvenile.

Experiment 2: metrifonate treatment

Adult animals were administered metrifonate (3–30 mg/kg, PO) 30 min before trial 1. Trial 2 was conducted at a retention time of 24 h, with either a familiar or an unfamiliar juvenile being placed into the observation cage. Placing an unfamiliar juvenile into the observation cage was used to reveal potential non-specific effects of the compound, which may influence investigation of the conspecific. Non-specific effects may result in a shorter investigation time of the unfamiliar juvenile. Compounds with specific effects on cognition are expected to have no effect on the duration of investigation for the unfamiliar juvenile.

Experiment 3: effects of scopolamine and metrifonate

Adult animals were injected with scopolamine (0.01–0.1 mg/kg, SC) 30 min before trial 1. Trial 2 was conducted at a retention time of 15 min, with the familiar juvenile. In an additional experiment, adult animals were injected with scopolamine (0.1 mg/kg, SC) and metrifonate (10–30 mg/kg, PO), both administered 30 min before trial 1. Trial 2 was conducted with the familiar juvenile after a retention time of 15 min.

Experiment 4: AR-R 17779 treatment

Adult animals were injected with AR-R 17779 (0.3–30 mg/kg, SC) 30 min before trial 1. Trial 2 was conducted with the familiar juvenile, after a retention time of 24 h. In an additional experiment, AR-R 17779 (10–30 mg/kg, SC) was injected immediately after trial 1. Trial 2 was conducted with the familiar juvenile after a retention time of 24 h. Finally, adult animals were injected with AR-R 17779 (1 mg/kg, SC) 30 min before trial 1, and trial 2 was conducted at a retention time of 24 h by placing an unfamiliar juvenile into the observation cage.

Experiment 5: effects of scopolamine and AR-R 17779

Adult animals were injected with scopolamine (0.1 mg/kg, SC) and AR-R 17779 (0.3–3 mg/kg, SC) 30 min before trial 1. Trial 2 was conducted at a retention time of 15 min, with the familiar juvenile being placed into the observation cage.

Experiment 6: effects of MLA and AR-R 17779

Adult animals were injected with AR-R17779 (1 mg/kg, SC) and MLA (10 µg, ICV) 30 min and 4 min before trial 1, respectively. Trial 2 was conducted with the familiar juvenile after a retention time of 24 h. In an additional experiment, adult animals were injected with MLA (10 µg, ICV) 4 min before testing. Trial 2 was conducted at a retention time of 15 min using the familiar juvenile.

Statistical analysis

Statistical analysis was performed using SAS 8.02 software (SAS Institute Inc. Cary, N.C., USA). The difference between social investigation at the two time points was used as a measure for social memory and calculated as percentage Reduction in Social Investigation Time (%RSIT)=[(SIT1−SIT2)/SIT1×100]. Dose-response data were subjected to a one-way ANOVA (factor: dose) and antagonist data were subjected to a two-way ANOVA (factor 1: scopolamine or MLA; factor 2: metrifonate or AR-R 17779). Post hoc comparisons were conducted using two-sample t-tests. All data were expressed as means+1 SEM. Statistical significance was set at P<0.05.

Results

Experiment 1: establishment of a forgetting curve

There was a significant main effect of delay interval [F(4,39)=9.556, P<0.001], thus supporting a decrease in %RSIT when the intervals between the first and the second encounter increase. Post hoc analysis revealed a significant decrease in %RSIT at 6 h and 24 h (P<0.01 and P<0.001, respectively) after trial 1, as compared with the 10-min interval (Fig. 1).
Fig. 1

Social recognition at different retention times. A reduction in investigation time during the second encounter is presented as percentage of investigation time measured during the first encounter. The second encounter was carried out after 10 min, 30 min, 1 h, 4 h, 6 h or 24 h, by placing the familiar juvenile in the test cage. * Indicates significant differences compared with the 10-min retention group: **P<0.01; ***P<0.001. Data are presented as means+1 SEM

Experiment 2: metrifonate treatment

A significant main effect of metrifonate treatment [F(3,36)=10.109, P<0.001] suggests a dose-dependent increase of %RSIT. At doses of 10 and 30 mg/kg, metrifonate induced a significant increase in %RSIT, employing a 24-h retention interval, as compared with the vehicle-treated group (P<0.001). Exposing adult drug-treated animals to an unfamiliar juvenile during the second encounter did not change the social interaction time (Fig. 2, inset; n=8 per group). No effects of metrifonate treatment have been seen on the investigation time during the first encounter (data not shown) (Fig. 2).
Fig. 2

Recognition-enhancing effects of metrifonate. Metrifonate (given 30 min prior to trial 1) was tested using a 24-h delay paradigm. The adult animal was exposed to the familiar juvenile as during the first encounter. The inset shows the effect of metrifonate when the adult animal was exposed to a different juvenile during the second encounter. n=10–14 per group. * Indicates significant differenes as compared to the vehicle-treated group: ***P<0.001. Data are presented as means+1 SEM

Experiment 3: effects of scopolamine and metrifonate

A significant main effect of scopolamine treatment [F(3,36)=7.683, P<0.001; Fig. 3a] was found. Post hoc analysis revealed that the 0.01 mg/kg and 0.1 mg/kg doses produced a significant decrease in %RSIT (P<0.01 and P<0.001, respectively). Scopolamine treatment did not show an effect on the exploration time during the first encounter (data not shown).

The disruptive effects of scopolamine were reversed by metrifonate (Fig. 3b). Two-way ANOVA revealed a significant main effect of metrifonate treatment [F(2,45)=12.45; P<0.001] and a significant interaction between scopolamine and metrifonate treatment [F(1,45)=14.98, P<0.001]. Post hoc analysis revealed that 1) scopolamine induced a marked decrease of %RSIT (P<0.001), suggesting that the adult rat did not recognize the juvenile which it had encountered 15 min earlier, and 2) that 10 mg/kg and 30 mg/kg metrifonate reversed this deficit (P<0.01 and P<0.001, respectively). Metrifonate given alone failed to increase %RSIT as compared with the vehicle-treated group (Fig. 3).
Fig. 3

Metrifonate reverses the effect of scopolamine. Scopolamine (given 30 min prior to trial 1) was tested using a 15-min paradigm (upper panel). The lower panel depicts the results obtained when metrifonate was co-administered with scopolamine (both given 30 min prior to trial 1). n=10 per group. * Indicates significant differences compared to the vehicle-treated group: *P<0.05; ***P<0.001. # Indicates significant differences compared to the scopolamine-treated group: #P<0.05; ##P<0.01; ###P<0.001. Data are presented as means+1 SEM

Experiment 4: AR-R 17779 treatment

AR-R 17779 treatment resulted in a significant main effect of dose [F(5,104)=3.55; P<0.01] and time [F(1,104)=14.20; P<0.001]. In the groups given AR-R 17779 prior to trial 1, post-hoc tests revealed significant increases in %RSIT at 1.0 mg/kg, 3.0 mg/kg, 10 mg/kg, and 30 mg/kg (P<0.01, P<0.001, P<0.01, and P<0.001, respectively). When given after trial 1, only the 30 mg/kg dose significantly increased %RSIT (P<0.05). When comparing corresponding doses between injections prior to and immediately after trial 1, significant differences were found at 10 mg/kg (P<0.05) and 30 mg/kg (P<0.001). Exposing adult animals treated prior to trial 1 with vehicle or AR-R 17779 to an unfamiliar juvenile, did not lead to an increase in %RSIT (Fig. 4, inset). AR-R 17779 treatment did not affect the investigation time during the first encounter (data not shown) (Fig. 4).
Fig. 4

Recognition-enhancing effects of AR-R 17779. AR-R 17779 was given immediately after trial 1 (filled squares) or 30 min prior to trial 1 (open squares). Twenty-four hours after the first encounter, the adult animal was exposed to the familiar juvenile during the second encounter. The inset shows the effect of AR-R 17779 (given 30 min prior to trial 1) when the adult animal was exposed to a different juvenile during the second encounter. Group sizes were: 0 mg/kg, pre-trial 1: n=20; post-trial 1: n=11; 0.3 mg/kg, pre-trial 1: n=10; 1.0 mg/kg, pre-trial 1: n=10; 3.0 mg/kg, pre-trial 1: n=20; 10 mg/kg, pre-trial 1: n=10; post-trial 1: n=11; 30 mg/kg, pre-trial 1: n=10; post-trial 1: n=11. * Indicates significant differences as compared to the vehicle-treated group: *P<0.05; **P<0.01; ***P<0.001. # Indicates significant differences as compared to the group given the same dose, but administered 30 min prior to trial 1: #P<0.05; ###P<0.001. Data are presented as means±1 SEM

Experiment 5: effects of scopolamine and AR-R 17779

A two-way ANOVA revealed significant main effects for scopolamine [F(1,103)=10.97; P<0.01] and for the interaction between AR-R 17779 and scopolamine [F(1,103)=4.54; P<0.05]. Post hoc analysis revealed that scopolamine at a dose of 0.1 mg/kg induced a marked decrease in %RSIT (P<0.001), employing a retention interval of 15 min. AR-R 17779 reversed this deficit at a dose of 1 mg/kg (P<0.01). AR-R 17779 given alone at a dose of 1 mg/kg did not further increase %RSIT compared with the vehicle-treated group. AR-R 17779 treatment did not have an effect on the investigation time during the first encounter, neither did scopolamine treatment (data not shown) (Fig. 5).
Fig. 5

AR-R 17779 reversal of the scopolamine-induced recognition deficit. AR-R 17779 was co-administered with scopolamine (both given 30 min prior to trial 1). After a 15-min interval, the adult animal was exposed to the familiar juvenile as during the first encounter. n=10–20 per group. * Indicates significant differences as compared to the vehicle-treated group: ***P<0.001. # Indicates significant differences as compared with the scopolamine-treated group: #P<0.05; ###P<0.001. Data are presented as means+1 SEM

Experiment 6: effects of MLA and AR-R 17779 (Fig. 6)

A two-way ANOVA revealed a significant main effect of AR-R 17779 [F(1,44)=6.49; P<0.05]. Post hoc t-tests revealed significant increases in %RSIT at 1 mg/kg (P<0.05). MLA (10 µg) reversed this increase (Fig. 6a; P<0.05). MLA given alone had no effect on %RSIT. However, at a retention time of 15 min, MLA significantly decreased %RSIT (P<0.05 compared to the vehicle-control group; Fig. 6b; n=9 per group). AR-R 17779 and MLA did not affect the investigation time during the first encounter (data not shown).
Fig. 6

MLA reversal of the recognition-enhancing effects of AR-R 17779. AR-R 17779 was given 30 min before MLA (10 µg ICV, given immediately prior to trial 1). After a 24-h interval the adult animal was exposed to the familiar juvenile as during the first encounter (upper panel). The lower panel shows the effect of MLA alone, employing a retention time of 15 min. n=8–16 per group. * Indicates significant differences as compared to the vehicle-treated group: *P<0.05. # Indicates significant differences as compared to the scopolamine-treated group: #P<0.05. Data are presented as means+1 SEM

Discussion

In the present study, the AchE-I metrifonate showed a behaviorally specific, cognitive-enhancing effect in both unimpaired rats, and a scopolamine-induced deficit model. These effects were mimicked by AR-R 17779, while the α7-nAChR antagonist, MLA, reversed the effects of AR-R 17779. This is to our knowledge the first study showing that α7-nAChRs mediate the cognitive-enhancing effects of AR-R 17779.

In a first series of experiments, we aimed to establish two conditions for subsequent testing of cognitive enhancers: a naturally forgetting model and a pharmacological deficit model. In general accordance with previous studies (Perio et al. 1989; Terranova et al. 1994; Popik and van Ree 1998), a retention interval of 10–60 min reduced the time spent by an old rat investigating a familiar juvenile for a second time. This effect was absent if the interval was increased to 4 h or more. From these studies, we selected a 24-h interval for assessing drug-induced enhancement of normal forgetting, and a 15-min interval for experiments with scopolamine. Variability in published forgetting curves may involve differences in housing conditions, as group-housed rats showed improved social memory in comparison with single housed rats (K. Selbach, unpublished observations). Likewise, group-housing has been reported to improve social memory in mice, which still recognized the familiar juvenile 24 h after the initial encounter; mice single-housed for 3 weeks or as short as 24 h failed to do so (Kogan et al. 2000). However, our own studies in group-housed mice revealed recognition of the familiar juvenile only up to 30 min after the initial trial (Schiegel, unpublished observations). The reason for this discrepancy is not known, but the housing studies illustrate the importance of procedural details in the expression of social memory. For establishment of a pharmacological deficit model, we selected the muscarinic antagonist scopolamine, since memory deficits induced by anticholinergics in healthy volunteers possess similarities to those seen in Alzheimer patients. Likewise, young monkeys treated with scopolamine show similar memory deficits as those observed in aged monkeys (Bartus 2000). In our hands, a robust disruption of social memory was found in rats at a dose of 0.1 mg/kg, SC, and therefore this dose was selected for reversal studies.

Further, we aimed to pharmacologically validate our model with the AChE-I metrifonate, and found that it was capable of improving social memory in unimpaired, as well as impaired (scopolamine-treated) rats. Since a 24-h retention interval was employed in the unimpaired rats and a 15-min interval in the impaired rats, metrifonate improved social memory over a wide time-interval, suggesting efficacy in ameliorating both short and long term memory. The effects were behaviorally specific, since metrifonate did not decrease social investigation time in the presence of an unfamiliar juvenile. Similarly, Gheusi et al. (1994) have shown cognitive-enhancing effects with the AChE-I tetrahydroaminoacridine in rats employing a 2-h retention interval. Behavioral specificity was supported by the absence of effects on social investigation time in the presence of a unfamiliar juvenile. Additionally, Winslow and Camarcho (1995) have shown reversal of a scopolamine-induced social recognition deficit with the AchE-I heptylphysostigmine. Together, these findings suggest that brain acetylcholine plays an important role in social memory acquisition and, possibly, consolidation.

In line with the well-established role of acetylcholine in cognitive functions, the memory-enhancing effects of nicotine and nicotinic agonists have been well described in rodents, non-human primates, and man across many different learning and memory tasks (e.g. Levin and Simon 1998), including the rat social recognition task (Perio et al. 1989). Different receptor subtypes may be involved in the effect of nicotine on learning and memory. Results from the present study support a role for α7-nAChR receptors in the social recognition test, as we found that AR-R 17779 improved social memory and that these effects were reversed by the α7-nAChR antagonist MLA. MLA itself induced a deficit in the social recognition task, which has also been described in the radial arm maze (Felix and Levin 1997; Bettany and Levin 2001). It cannot be completely ruled out that the detrimental effect of MLA on social recognition contributed to the reversal of the effects of AR-R 17779, although the MLA-induced deficit in social recognition was obtained at a 15-min retention interval, whereas the reversal study employed a 24-h interval, and the MLA-vehicle treated group did not show a decrease in social recognition at this interval (Fig. 6a). In addition to the current activity in the social recognition task, AR-R 17779 reversed the impairment in working memory performance induced by fimbria-fornix lesion in a radial arm maze task (Levin et al. 1999), and improved object memory in an object recognition task (A. Blokland, unpublished observations). AR-R 17779 was given pre-training in these studies and showed activity at comparable doses in the social recognition, object recognition, and radial arm maze tasks (minimal effective dose: 1, 3, and 1 mg/kg, respectively).

Although the data obtained with AR-R 17779 in the social recognition, object recognition, and radial arm maze task provide good evidence for a role of α7-nACh receptors in learning and memory, these effects appear not to extend across all learning and memory tasks. AR-R 17779 failed to reverse a scopolamine-induced deficit in spatial memory in a Morris water maze task (van der Staay, unpublished observations). In an attention task, (five-choice-serial-reaction-time-task: 5-CSRTT), AR-R 17779 and MLA had no effect. In contrast, the α4β2-nAChR agonist SIB1765F increased correct responding (Grottick and Higgins 2000; Hahn et al. 2001). Additionally, the effects of nicotinic drugs in the 5-CSRTT were antagonized by an α4β2 antagonist (dihydroxy-beta-erythroidine-DhβE), but not by an α7-nAChR antagonist (MLA). Although these results suggest that effects of nicotinic drugs in this model involve activation of α4β2-nACh receptors (Grottick and Higgins 2000), the role of the α7-nAChR in the 5-CSRTT needs to be further addressed by studies with AR-R 17779 employing different doses and different treatment durations. The present pattern of action in attention tests (i.e. a role for α4β2-nAChRs but not for α7-nAChRs) is reminiscent of the differential effects found with α7 and α4β2-nACh receptor agonists in a model of pre-attentive processes, i.e. the mouse prepulse inhibition paradigm. In this model, the α7-nAChR agonists AR-R 17779 and GTS-21 were inactive, whereas the α4β2-nAChR agonist A-85380 augmented sensorimotor gating (Schreiber et al. 2002). Although these data do not support a role for α7-nAChRs in processes of attention, the recent observation that GTS-21 improved attention in humans (Kitagawa et al. 2003) suggests that the role of α7-nAChR in attentional processes needs to be re-addressed in further models.

In the present study, the AchE-I metrifonate and the α7-nAChR agonist AR-R 17779 decreased %SIT. These results suggest that both drugs improve social memory, but the potential involvement of alternative behavioral mechanisms, such as, for example, drug-induced conditioned taste aversion, cannot be ruled out. Studies with MLA suggest that the effects of AR-R 17779 are mediated by activation of α7-nACh receptors. Interestingly, AR-R 17779 was clearly more potent in improving social memory when given before rather than after the first exposure. The predominant effect on the acquisition (and, possibly, early consolidation) phase of the memory process in the SRT is consistent with previous findings of Levin et al. (1999), who tested AR-R 17779 in the radial arm maze where it improved repeated acquisition. Further studies employing different treatment schedules and taking into account the pharmacokinetic properties of AR-R 17779 need to be performed to further confirm and extend these findings.

The potent and robust activity of AR-R 17779 in improving baseline social memory and reversing a scopolamine-induced deficit suggests that α7-nACh receptor agonists may constitute of a novel class of cognitive-enhancers with potential efficacy against disturbances in short- and long term memory. Indeed, phase I studies in healthy volunteers with GTS-21 reported an improvement of performance in several tests of a CDR computerized test battery, including tests for working memory, long term memory, and attention (Kitagawa et al. 1998). The finding with GTS-21 on attention appears not consistent with the inactivity of AR-R 17779 in the 5-CSRTT task (Grottick and Higgins 2000), and future studies need to address whether the 5-CSRTT lacks predictive validity for attention-improving effects of α7-nAChR agonists or whether differences in the pharmacological properties of both α7-nAChR agonists underlie their different effects on attention. Altogether, the present data support that α7-nAChR agonists could prove useful for the treatment of cognitive deficits in Alzheimer’s disease, schizophrenia, and other neurological and psychiatric disorders. Results from phase II studies are eagerly awaited to further substantiate the therapeutic potential of α7-nACh receptor agonists.

Acknowledgements

Dr. Jordan Mechanic is thanked for his critical review of the manuscript and helpful comments. Dr. van der Staay is thanked for the helpful discussions.

Copyright information

© Springer-Verlag 2004